Skip to content

Translation in progress Some entries still include text in the original language.

Oct 12 · Monday

All papers

Found 6 papers
Filter4

Impact

Harmful content16Unauthorized actions47Data and privacy leaks20Output manipulation35Availability and reliability6Oversight failure39

System type

AI Agent4

Layer

Application layer3

System component

MCP and skills5Memory3Guardrails and judges5Monitors1Reasoning chains4Vision5Audio3Video2Multimodal10
Select any value within a category; different categories must all match. Counts include the other filters you have selected.Another 254papers have no tags yet and are not included in filtered results.
Impact：Availability and reliabilitySystem component：MemorySystem component：MonitorsSystem component：VideoClear all

Another 254papers do not have category tags yet and are not included in filtered results.

  3. OtherarXiv 2609.08832

    自进化智能体框架缩小 LLM 智能体的「一致性差距」

    提出离线两阶段框架，诊断易翻转步骤并生成记忆指南以稳定智能体

    arXiv
    2609.08832
    Published
    Layer
    Application layer
    Setting
    AI Agent
    Evaluated models
    GPT-4.1, GPT-OSS-120B
    Abstract用离线一致性诊断生成记忆指南，在 AppWorld 上提高重复全通过率。

    作者要缩小的是 LLM 智能体平均通过、重复不稳之间的 consistency gap，即 Mean@k 减 Passˆk。

    Paper analysisRead analysis
  4. Evaluation and benchmarksarXiv 2608.18066

    Salesforce AI Research 重测自改进 Agent

    重测记忆型自改进 Agent 的运行方差与任务顺序敏感性

    arXiv
    2608.18066
    Published
    Layer
    Application layer
    Setting
    AI Agent
    Evaluated models
    GPT-5-mini, Gemini-2.5-Pro, GPT-OSS-120B and others4
    Abstract论文揭示了基于记忆的自改进智能体在多次运行和乱序下的脆弱性，并指出规约不完备是诱发退化的关键瓶颈。

    这篇论文是对基于文本记忆的自我改进智能体在复杂环境中评估可靠性的实证研究。

    Paper analysisRead analysis
  5. AttacksarXiv 2605.30040

    研究揭示 LLM 按 token 计费可被服务商系统性虚报

    构造可通过计费审计的 token 虚报，使服务商系统性超收

    arXiv
    2605.30040
    Published
    Layer
    Infrastructure layer
    Setting
    Models and APIs
    Evaluated models
    PALACE domain-adapted auditor
    Abstract不诚实提供方可在三种 token 审计上虚报并通过检查，作者认为需改用提供方不控制的证据。

    不透明 LLM 的按 token 计费里，现有审计核验的是提供方可控的证据，不诚实提供方可以虚报并通过检查。。

    Paper analysisRead analysis
  6. Evaluation and benchmarksarXiv 2602.10639

    VideoSTF 揭示视频大模型输出重复问题

    提出 VIDEO STF 评测视频大模型输出重复并压力测试时序扰动

    arXiv
    2602.10639
    Published
    Layer
    Prompt layer
    Setting
    Models and APIs
    Evaluated models
    LLaVA-Video-7B-Qwen2, LLaVA-Video-7B-Qwen2-Video-Only, LLaVA-NeXT-Video-7B and others10
    AbstractVIDEO STF 把输出重复做成可测、可被时间扰动放大、并可黑盒诱导的失败。

    VIDEO STF 测量 VideoLLM 的输出重复，并把时间变换用作 black-box 诱导。

    Paper analysisRead analysis
You have reached the end